chroma-core/chroma

Health v2 · maintenance only

Adoption not part of the score

29152 stars · 2463 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Chroma is an open-source search and retrieval database for AI applications, supporting dense vector, sparse (BM25/SPLADE), full-text, regex, and metadata search with built-in embedding storage. It can be run locally, self-hosted, or used via the managed serverless Chroma Cloud, with SDKs for Python, TypeScript, and Rust.

Use cases

• store and query embeddings for a RAG chatbot
• build semantic search over documents with metadata filtering
• add hybrid vector and keyword search to an AI app
• index a codebase for a coding agent to search
• prototype an in-memory vector store then persist it
• build agentic search that iteratively refines queries
• self-host a vector database for LLM applications

When to choose

• you need an easy-to-start open-source vector database for RAG or AI agents
• you want hybrid dense/sparse/full-text search with metadata filtering in one system
• you want a simple API with Python, TypeScript, or Rust clients
• you want a managed serverless option without ops overhead

When to avoid

• you need a general-purpose relational database with joins and transactions
• you need strict SQL or graph query capabilities
• your workload is small enough that a simple in-process embedding index suffices
• you require a fully offline solution but rely on Chroma Cloud features

Facets

service · maturity active

vector-database search-engine database rag databases artificial-intelligence large-language-models python rust self-hosted cloud cross-platform vector-database embeddings semantic-search hybrid-search full-text-search rag ai-infrastructure chroma-cloud open-source search retrieval-augmented-generation ai-agents nodejs docker

Member repositories

For agents

markdown · JSON · MCP: product_card(name="chroma-core/chroma")